On Saturday, May 18, 2002, at 10:28 PM, Autrijus Tang wrote: > Hi all. I'm considering a new feature of cpansmoke 0.06, which > involves the > typical FAIL report, > ... > ...it is possible to [re-run failed tests], attaching the output to > the cpansmoke report.
> 1) Is such a feature worth implementing? Yes, indeed! > 2) Should it be a switch of cpansmoke, or comes by default, or as > an additional > option in "Send test results [Y/n]?"? My first choice is to make it a default, but making it an option would be just fine as well. The more info an author gets about failures, the more able that author will be to address them. > 3) Should we limit the size of report it generates? Yes, but I'm not sure what a good size limit would be... If a report exceeds the limit, I suggest truncating it, and appending a note saying it was truncated -- e.g. "Test output truncated by CPANPLUS. Contact the author of this message for more detail, if needed." Then send the output that came before the limit was reached. Keep it up! This is great... - Brian.
